BACKGROUND

Modem vehicles are often equipped with brake signal lights, e.g., two conventional brake signal lights at both ends of the rear and a high-mounted brake signal light in the upper middle of the rear. When a vehicle is performing braking or emergency braking, e.g., when a brake pedal of the vehicle is depressed, these brake signal lights will illuminate to remind its rear vehicle to take a braking operation as soon as possible, thereby reducing the risk of collision or rear-end collision.

SUMMARY

This Summary is provided to introduce a selection of concepts that are further described below in the Detailed Description. It is not intended to identify key features or essential features of the claimed subj ect matter, nor is it intended to be used to limit the scope of the claimed subj ect matter. Embodiments of the present disclosure propose a method, apparatus and computer program products for brake alerting based on context detection. It may be detected whether a front vehicle in front of a current vehicle is braking or is about to brake. In response to detecting that the front vehicle is braking or is about to brake, a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle may be predicted based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake. It may be determined whether the rear collision probability is greater than a rear collision threshold. In response to determining that the rear collision probability is greater than the rear collision threshold, a rear vehicle brake alert may be issued to the rear vehicle.

The embodiments of the present disclosure also propose a system for brake alerting based on context detection. The system may comprise: a processing unit, configured to: detect whether a front vehicle in front of a current vehicle is braking or is about to brake, in response to detecting that the front vehicle is braking or is about to brake, predict a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake, determine whether the rear collision probability is greater than a rear collision threshold, and in response to determining that the rear collision probability is greater than the rear collision threshold, send a rear vehicle brake alerting command to a rear vehicle brake alerting unit; and a rear vehicle brake alerting unit, configured to: in response to receiving the rear vehicle brake alerting command from the processing unit, issuing a rear vehicle brake alert to the rear vehicle.

It should be noted that the above one or more aspects comprise the features hereinafter fully described and particularly pointed out in the claims. The following description and the drawings set forth in detail certain illustrative features of the one or more aspects. These features are only indicative of the various ways in which the principles of various aspects may be employed, and this disclosure is intended to include all such aspects and their equivalents.

BRIEF DESCRIPTION OF THE DRAWINGS

The disclosed aspects will hereinafter be described in connection with the appended drawings that are provided to illustrate and not to limit the disclosed aspects.

FIG.l illustrates an exemplary architecture of a system for brake alerting based on context detection according to an embodiment of the present disclosure.

FIG.2 illustrates an exemplary process for brake alerting based on context detection according to an embodiment of the present disclosure.

FIG.3 illustrates an exemplary process for calculating an absolute speed of a front vehicle according to an embodiment of the present disclosure.

FIG.4 illustrates an exemplary process for predicting a rear collision probability of a rear vehicle colliding with a current vehicle according to an embodiment of the present disclosure.

FIGs.5A to 5G illustrate examples of brake alerting based on context detection according to embodiments of the present disclosure.

FIG.6 is a flowchart of an exemplary method for brake alerting based on context detection according to an embodiment of the present disclosure.

FIG.7 illustrates an exemplary system for brake alerting based on context detection according to an embodiment of the present disclosure.

FIG.8 illustrates an exemplary apparatus for brake alerting based on context detection according to an embodiment of the present disclosure.

DETAILED DESCRIPTION

The present disclosure will now be discussed with reference to several example implementations. It is to be understood that these implementations are discussed only for enabling those skilled in the art to better understand and thus implement the embodiments of the present disclosure, rather than suggesting any limitations on the scope of the present disclosure.

At present, a driver of a rear vehicle, while driving, may determine whether a current vehicle in front of the rear vehicle is braking through observing a brake signal light on the current vehicle. Herein, vehicles that are in the same lane but located in front of and behind each other may be referred to by a rear vehicle, a current vehicle, and a front vehicle. A rear vehicle is behind a current vehicle. A front vehicle is in front of the current vehicle. If a driver of a rear vehicle determines that a current vehicle is braking, the driver may take a braking operation accordingly, so as to try his best to avoid the rear vehicle colliding or rear-ending the current vehicle. If a sufficient safe distance may be kept between the rear vehicle and the current vehicle, the road surface is in good condition, and the driver of the rear vehicle is able to take the braking operation within a reasonable response time after observing the braking of the current vehicle, a rear-end collision between the rear vehicle and the current vehicle is usually avoidable. However, it is sometimes difficult to guarantee the safe distance between vehicles on a road with heavy traffic and high traffic flow. Additionally, when the road surface becomes slippery due to weather such as rain, snow, etc., the braking distance of the vehicle will increase. Furthermore, the response time for the driver of the rear vehicle to take a braking operation may increase due to distraction or other reasons. Various factors, such as the factors described above, cause collisions or rear-end collisions on actual roads to occur from time to time.

Typically, a current vehicle braking is because a driver of the current vehicle has taken a braking operation as the driver of the current vehicle observes that a front vehicle is braking. While the current vehicle starts to brake, i.e., while its brake pedal is depressed, a brake signal light of the current vehicle are usually illuminated, so that a driver of a rear vehicle may observe that the current vehicle is braking. If the driver of the rear vehicle may know in advance, e.g., before the current vehicle has not yet braked, that the current vehicle is about to brake, the driver of the rear vehicle may take the braking operation earlier to gain more response time. There is a solution that makes the driver of the rear vehicle to know in advance that the current vehicle is about to brake. The solution is that when the current vehicle detects that a brake signal light of the front vehicle illuminates, the current vehicle also illuminates its own brake signal light, thereby passing the brake signal of the front vehicle to the rear vehicle. However, in many cases, the brake signal of the front vehicle is meaningless to the rear vehicle. If all brake signals of the front vehicle are passed to the rear vehicle, it will cause disturbance to the rear vehicle. For example, when the rear vehicle is far away from the current vehicle, even if the current vehicle is braking correspondingly due to the emergency braking of the front vehicle, the rear vehicle may not need to take a braking operation immediately.

Embodiments of the present disclosure propose brake alerting based on context detection. When a current vehicle detects that a front vehicle is braking or is about to brake, it may determine whether to issue a rear vehicle brake alert to a rear vehicle based on a rear context of the current vehicle. Herein, information related to a rear environment of a current vehicle may be referred to as a rear context of the current vehicle. The rear context of the current vehicle may include, e.g., a distance between a rear vehicle and the current vehicle, a relative speed of the rear vehicle relative to the current vehicle, a road condition, a weather condition, a vehicle type of the rear vehicle, etc. The rear vehicle brake alert may alert the rear vehicle to take a braking operation. When detecting that the front vehicle is braking or is about to brake, a rear collision probability of the rear vehicle colliding with the current vehicle may be predicted based on the rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake. Herein, a probability of a rear vehicle colliding with a current vehicle may be referred to as a rear collision probability. If the rear collision probability is greater than a predetermined rear collision threshold, i.e., if the rear vehicle is likely to collide or rear-end the current vehicle, a rear vehicle brake alert may be issued to the rear vehicle; and if the rear collision probability is not greater than the predetermined rear collision threshold, i.e., if the rear vehicle is not likely to collide or rear-end the current vehicle, a rear vehicle brake alert will not be issued to the rear vehicle. That is, the rear vehicle brake alert may be issued to the rear vehicle only when the rear vehicle is likely to collide or rear-end the current vehicle. In the above manner, an accurate rear vehicle brake alert may be passed to a rear vehicle. Additionally, if each vehicle may determine, based on its rear context, whether or not to issue a rear vehicle brake alert to a vehicle behind it, accurate rear vehicle brake alerts may be passed in sequence to the rear, thereby significantly reducing the risk of collisions or rear-end collisions on roads.

In an aspect, the embodiments of the present disclosure propose to detect whether a front vehicle is braking or is about to brake through a plurality of approaches. In an implementation, it may be detected whether a front vehicle is braking or is about to brake through detecting whether the front vehicle issues a brake alert. For example, it may be detected whether the front vehicle issues a visual signal, an infrared signal, an acoustic signal and a wireless signal, etc., for alerting the current vehicle to brake. In another implementation, it may be detected whether a front vehicle is braking through detecting whether the drop in an absolute speed of the front vehicle exceeds a drop threshold. When it is detected that the drop in the absolute speed of the front vehicle exceeds the drop threshold, it may be determined that the front vehicle is braking. In particular, when it is detected that the drop in the absolute speed of the front vehicle exceeds the drop threshold within a very short period of time, it may be determined that the front vehicle has undergone an emergency braking. It may be detected whether a front vehicle is braking or is about to brake through either of the two implementations described above. In this manner, it may be detected in time and accurately whether the front vehicle is braking or is about to brake, which facilitates the current vehicle to issue an accurate rear vehicle brake alert to a rear vehicle in time and take a braking operation in time.

In another aspect, the embodiments of the present disclosure propose to issue, to a rear vehicle, a rear vehicle brake alert. For example, a rear vehicle brake alert may be issued to a rear vehicle through sending at least one of a visual signal, an infrared signal, an acoustic signal and a wireless signal for alerting the rear vehicle to brake. These signals may be issued through corresponding hardware unit of the current vehicle. For example, the visual signal may be issued through a brake signal light of the current vehicle. The brake signal light may be existing lights on the vehicle for indicating that the brake pedal is depressed, e.g., conventional brake signal lights at both ends of the rear or a high-mounted brake signal light in the upper middle of the rear. According to the embodiments of the present disclosure, a brake signal light on a vehicle may also be used to issue a rear vehicle brake alert to a vehicle behind it. Preferably, the visual signal may be issued through a brake alerting light of the current vehicle. The brake alerting light may be a light added at the rear of the vehicle, and specially configured to issue a rear vehicle brake alert to a vehicle behind it. The shape of the brake alerting light, or the characteristics of the light it emits, may be significantly different from the brake signal light, so that the rear vehicle brake alert it emits may be easily detected by the rear vehicle.

In yet another aspect, the embodiments of the present disclosure propose when it is detected that a front vehicle is braking or is about to brake, determining whether to send a current vehicle braking command within a current vehicle based on a front context of the current vehicle. Herein, information related to a front environment of a current vehicle may be referred to as a front context of the current vehicle. The front context of the current vehicle may include, e.g., a distance between a front vehicle and the current vehicle, a relative speed of the front vehicle relative to the current vehicle, a road condition, a weather condition, a vehicle type of the front vehicle, etc. When detecting that the front vehicle is braking or is about to brake, a front collision probability of a current vehicle colliding with the front vehicle may be predicted based on the front context of the current vehicle and an assumption that the front vehicle and the current vehicle brake. Herein, a probability of a current vehicle colliding with a front vehicle may be referred to as a front collision probability. If the front collision probability is greater than a predetermined front collision threshold, i.e., if the current vehicle is likely to collide or rear-end the front vehicle, a current vehicle braking command may be sent within the current vehicle; and if the front collision probability is not greater than the predetermined front collision threshold, i.e., if the current vehicle is not likely to collide or rear-end the front vehicle, a current vehicle braking command may not be sent within the current vehicle. Sending the current vehicle braking command within the current vehicle may prompt a driver of the current vehicle or an automatic braking unit to take a braking operation, so as to avoid the current vehicle colliding or rear-ending a front vehicle. If the current vehicle braking command is not sent within the current vehicle, the current vehicle may continue to drive forward for a certain distance, and the current vehicle braking command may be sent within the current vehicle only when the front collision probability is greater than the front collision threshold. If at this point, a rear vehicle may collide or rear-end the current vehicle, and the rear vehicle is able to immediately perform emergency braking when receiving a rear vehicle brake alert, and the current vehicle continues to drive for a certain distance on the premise of ensuring that it will not rear-end the front vehicle, the distance between the rear vehicle and the current vehicle may be increased. In this way, the risk of the rear vehicle rear-ending the current vehicle may be reduced.

In still another aspect, the embodiments of the present disclosure propose to determine how to send a current vehicle braking command within a current vehicle according to a vehicle type of the current vehicle. If the current vehicle is a human-driven vehicle, the current vehicle brake command may be sent to a current vehicle brake alerting unit in the current vehicle, so that the current vehicle brake alerting unit issues a current vehicle brake alert. The current vehicle brake alert may alert a driver of the current vehicle to implement a braking operation as soon as possible, such as an emergency braking operation. Preferably, the braking degree of the current vehicle brake alert issued by the current vehicle braking unit may be associated with a front collision probability. If the current vehicle is an autonomous vehicle, the current vehicle braking command may be sent at least to an automatic braking unit in the current vehicle, so that the automatic braking unit implements an automatic braking operation. Preferably, the braking degree of the automatic braking operation implemented by the automatic braking unit may be associated with a front collision probability.

FIG.l illustrates an exemplary architecture 100 of a system for brake alerting based on context detection according to an embodiment of the present disclosure. The system may be deployed on any vehicle, e.g., a human-driven vehicle, an autonomous vehicle, etc.

The architecture 100 may include at least one receiving unit 110. The at least one receiving unit 110 may receive at least front information 102 from the front of a current vehicle and rear information 104 from the rear of the current vehicle. The front information 102 may include information of a front vehicle located in front of the current vehicle. The rear information 104 may include information of a rear vehicle located behind the current vehicle. In an implementation, the at least one receiving unit 110 may include a front camera facing the front of the current vehicle and a rear camera facing the rear of the current vehicle. The front camera may receive the front information 102. The rear camera may receive the rear information 104. In another implementation, the at least one receiving unit 110 may include a LiDAR. The LiDAR may receive information around the current vehicle, including the front information 102 and the rear information 104.

The at least one receiving unit 110 may process the front information 102 and the rear information 104, and generate a front input 112 and a rear input 114 suitable for being processed by a processing unit 120. When the at least one receiving unit 110 includes a front camera and a rear camera, the front input 112 and the rear input 114 may be an image corresponding to a front environment of the current vehicle and an image corresponding to a rear environment of the current vehicle, respectively. For example, the front input 112 may contain an image of a front vehicle, and the rear input 114 may contain an image of a rear vehicle. When the at least one receiving unit 110 includes a LiDAR, the front input 112 and the rear input 114 may be a point cloud corresponding to the front environment of the current vehicle and a point cloud corresponding to the rear environment of the current vehicle, respectively. For example, the front input 112 may contain a point cloud of the front vehicle, and the rear input 114 may contain a point cloud of the rear vehicle.

The processing unit 120 may be a chipset system including an artificial intelligence chip. The processing unit 120 may perform a process for brake alerting based on context detection based on the front input 112 and the rear input 114. For example, the processing unit 120 may detect whether the front vehicle is braking or is about to brake. If it is detected that the front vehicle is braking or is about to brake, the processing unit 120 may predict a rear collision probability of a rear vehicle colliding with the current vehicle based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake. If the rear collision probability is greater than a rear collision threshold, it means that there is a high probability of the rear vehicle colliding with the current vehicle, i.e., the rear vehicle will most likely rear-end the current vehicle. In this case, the processing unit 120 may issue a rear vehicle brake alert 132 to the rear vehicle for alerting the rear vehicle to brake, so as to alert the rear vehicle to brake as soon as possible. For example, the processing unit 120 may generate a rear vehicle brake alerting command 122, and send it to a rear vehicle brake alerting unit 130. The rear vehicle brake alerting unit 130 may receive the vehicle brake alerting command 122, and issue a rear vehicle brake alert 132 to the rear vehicle. The rear vehicle brake alerting unit 130 may be a hardware unit capable of issuing at least one of a visual signal, an infrared signal, an acoustic signal and a wireless signal. Accordingly, the rear vehicle brake alert 132 may include, e.g., a visual signal, an infrared signal, an acoustic signal, a wireless signal, etc., for alerting the rear vehicle to brake. As an example, the rear vehicle brake alerting unit 130 may be a brake signal light of the current vehicle. The brake signal light may be existing lights on the vehicle for indicating that the brake pedal is depressed, e.g., conventional brake signal lights at both ends of the rear or a high-mounted brake signal light in the upper middle of the rear. According to the embodiments of the present disclosure, a brake signal light on a vehicle may also be used to issue the rear vehicle brake alert 132 to a vehicle behind it. Preferably, the rear vehicle brake alerting unit 130 may be issued through a brake alerting light of the current vehicle. The brake alerting light may be a light added at the rear of the vehicle, and specially configured to issue a rear vehicle brake alert to a vehicle behind it. The shape of the brake alerting light, or the characteristics of the light it emits, may be significantly different from the brake signal light, so that the rear vehicle brake alert 132 it emits may be easily detected by the rear vehicle.

Additionally, if it is detected that the front vehicle is braking or is about to brake, the processing unit 120 may predict a front collision probability of the current vehicle colliding with the front vehicle based on a front context of the current vehicle and an assumption that the front vehicle and the current vehicle brake. If the front collision probability is greater than a front collision threshold, it means that there is a high probability of the current vehicle colliding with the front vehicle, i.e., the current vehicle will most likely rear-end the front vehicle. In this case, the processing unit 120 may send a current vehicle braking command 124 within the current vehicle. If the current vehicle is a human-driven vehicle, the processing unit 120 may send a current vehicle brake command 124 to a current vehicle brake alerting unit 140 in the current vehicle, so that the current vehicle brake alerting unit 140 issues a current vehicle brake alert 142. The current vehicle brake alerting unit 140 may be a hardware unit capable of issuing at least one of a visual signal, an acoustic signal, and a haptic signal. Accordingly, the current vehicle brake alert 142 may include, e.g., a visual signal, an acoustic signal, and a haptic signal, etc. The current vehicle brake alert 142 may alert a driver of the current vehicle to implement a braking operation as soon as possible. Preferably, the braking degree of the current vehicle brake alert 142 issued by the current vehicle braking unit 140 may be associated with the front collision probability. The greater the front collision probability, the greater the intensity of the current vehicle brake alert 142 may be, so as to cause the driver to perform a braking operation with a greater braking degree. For example, when the current vehicle brake alert 142 is an acoustic signal, if the front collision probability is a large value greater than the front collision threshold, the acoustic signal may be urgent; and if the front collision probability is a small value greater than the front collision threshold, the acoustic signal may be flat.

If the current vehicle is an autonomous vehicle, the processing unit 120 may send a current vehicle braking command 124 at least to an automatic braking unit 150 in the current vehicle, so that the automatic braking unit 150 implements an automatic braking operation 152. Preferably, the braking degree of the automatic braking operation 152 implemented by the automatic braking unit 150 may be associated with the front collision probability. The greater the front collision probability, the greater the braking degree of the automatic braking operation 152 may be, so as to cause the current vehicle to stop as soon as possible. Optionally, if the current vehicle is the autonomous vehicle, the processing unit 120 may also send a current vehicle braking command to the current vehicle brake alerting unit 140 in the current vehicle, so that the driver in the current vehicle may also know that the current vehicle is about to be automatically braked, and may take over the vehicle and take appropriate actions if necessary. That is, if the current vehicle is the autonomous vehicle, the architecture 100 may include only the automatic braking unit 150 or include both the current vehicle brake alerting unit 140 and the automatic braking unit 150.

An exemplary process performed by the processing unit 120 for brake alerting based on context detection will be described later in conjunction with FIG.2.

At least some units of the units described above in the architecture 100 may be obtained through improving existing components on the vehicle. For example, a modern vehicle is usually equipped with a front camera and a rear camera. These two cameras may be used as the receiving unit 110 in the architecture 100. Additionally, the modem vehicle is usually equipped with chipset systems, such as chipset systems used for Automatic Parking Assist System (APAS), Front Looking System (FLS), car entertainment systems, etc. Such chipset systems may be improved in accordance with the embodiments of the present disclosure, to be used as the processing unit 120 in the architecture 100. Therefore, it is possible to enable the vehicle to implement brake alerting based on context detection at a low cost without the need to significantly change the hardware structure of the vehicle.

It should be appreciated that the architecture shown in FIG.l is only one example of the architecture of the system for brake alerting based on context detection. Depending on actual application requirements, the system for brake alerting based on context detection may have any other architecture, and may include more or fewer elements. Additionally, the various units in the architecture 100 may have any other form of implementation. For example, the receiving unit 110 may also include a sensor, such as a millimeter wave radar, in addition to being a camera and a LiDAR.

FIG.2 illustrates an exemplary process 200 for brake alerting based on context detection according to an embodiment of the present disclosure. The process 200 may be performed, e.g., through the processing unit 120 in FIG.l.

At 202, a front input may be obtained. The front input may be received, e.g., from a receiving unit. In the case where the receiving unit is a camera, the front input may be an image corresponding to an environment in front of a current vehicle. The image may include an image of a front vehicle in front of the current vehicle. In the case where the receiving unit is a LiDAR, the front input may be a point cloud corresponding to the environment in front of the current vehicle. The point cloud may include a point cloud of the front vehicle.

At 204, it may be detected, based on the front input, whether the front vehicle is braking or is about to brake.

In an implementation, it may be detected whether the front vehicle is braking or is about to brake through detecting whether the front vehicle issues a brake alert. For example, it may be detected whether the front vehicle issues a visual signal, an infrared signal, an acoustic signal, a wireless signal, etc., for alerting the current vehicle to brake. The visual signal, the infrared signal, the acoustic signal and the wireless signal, etc., may be sent through corresponding hardware units of the front vehicle. As an example, the visual signal may be issued through a brake signal light or a brake alerting light of the front vehicle. The brake signal light may be existing lights on the vehicle for indicating that the brake pedal is depressed, e.g., conventional brake signal lights at both ends of the rear or a high-mounted brake signal light in the upper middle of the rear. According to the embodiments of the present disclosure, a brake signal light on a vehicle may also be used to issue a vehicle brake alert to a vehicle behind it. The brake alerting light may be a light added at the rear of the vehicle, and specially configured to issue a vehicle brake alert to a vehicle behind it. For example, when a front vehicle observes an emergency situation in front of it, it may perform emergency braking immediately, and at the same time alert a vehicle behind it to take a braking operation through the brake alerting light. According to the form of the brake alert issued by the front vehicle and the form of the receiving unit of the current vehicle, it may be detected whether the front vehicle issues a brake alert by employing a corresponding method. For example, when the brake alert is a visual signal issued by a brake signal light or a brake alerting light, and the receiving unit of the current vehicle is a camera, it may be determined whether the brake signal light or the brake alerting light illuminates through performing image detection on the image provided by the camera, thereby detecting whether the front vehicle issues the brake alert.

In another implementation, it may be detected whether the front vehicle is braking through detecting whether the drop in an absolute speed of the front vehicle exceeds a drop threshold. When it is detected that the drop in the absolute speed of the front vehicle exceeds the drop threshold, it may be determined that the front vehicle is braking. In particular, when it is detected that the drop in the absolute speed of the front vehicle exceeds the drop threshold within a very short period of time, it may be determined that the front vehicle has undergone an emergency braking. FIG.3 illustrates an exemplary process 300 for calculating an absolute speed of a front vehicle according to an embodiment of the present disclosure. At 302, a change in a distance between a front vehicle and a current vehicle over a predetermined time may be estimated. The distance between the front vehicle and the current vehicle may be estimated through a known inter-vehicle ranging technology. At 304, a relative speed of the front vehicle relative to the current vehicle may be calculated based on the change and the predetermined time. At 306, an absolute speed of the front vehicle may be calculated based on the relative speed of the front vehicle relative to the current vehicle and an absolute speed of the current vehicle.

It may be detected whether the front vehicle is braking or is about to brake through either of the two implementations described above. In this manner, it may be detected in time and accurately whether the front vehicle is braking or is about to brake, which facilitates the current vehicle to issue an accurate rear vehicle brake alert to a rear vehicle in time and take a braking operation in time.

If at 204 it is not detected that the front vehicle is braking or is about to brake, the process 200 may return to 202, where a front input continues to be obtained.

If at 204 it is detected that the front vehicle is braking or is about to brake, the process 200 may proceed to 206 and 216 synchronously.

At 206, a rear input may be obtained. The rear input may be received, e.g., from a receiving unit. In the case where the receiving unit is a camera, the rear input may be an image corresponding to an environment behind the current vehicle. The image may include an image of a rear vehicle behind the current vehicle. In the case where the receiving unit is a LiDAR, the rear input may be a point cloud corresponding to the environment behind the current vehicle. The point cloud may include a point cloud of the rear vehicle.

At 208, a rear context of the current vehicle may be obtained through analyzing the rear input. The rear context may include, e.g., a distance between the rear vehicle and the current vehicle, a relative speed of the rear vehicle relative to the current vehicle, a road condition, a weather condition, a vehicle type of the rear vehicle, etc. The rear context may be obtained in a known way. For example, the distance between the rear vehicle and the current vehicle may be obtained through a known inter-vehicle ranging technology. For a relative speed of the rear vehicle relative to the current vehicle, a change of the distance between the rear vehicle and the current vehicle within a predetermined time may be estimated first, and then the relative speed of the rear vehicle relative to the current vehicle may be calculated based on the change and the predetermined time. The road condition, the weather condition, the vehicle type of the rear vehicle, etc., may be obtained through a known image classification technology.

At 210, a rear collision probability of the rear vehicle behind the current vehicle colliding with the current vehicle may be predicted based on the rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake. The rear collision probability may be normalized to a number within a range [0, 1], A greater rear collision probability means that there is a higher probability of the rear vehicle colliding with the current vehicle, i.e., the rear vehicle will more likely rear-end the current vehicle. An exemplary process for predicting a rear collision probability will be described later in conjunction with FIG.4.

At 212, it may be determined whether the rear collision probability is greater than a rear collision threshold.

If it is determined that the rear collision probability is greater than the rear collision threshold, the process 200 may proceed to 214. At 214, a rear vehicle brake alert may be issued to the rear vehicle. For example, the rear vehicle brake alerting command may be sent to a rear vehicle brake alerting unit in the current vehicle, so that the rear vehicle brake alerting unit issues a rear vehicle brake alert. The rear vehicle brake alert may include, e.g., a visual signal, an infrared signal, an acoustic signal, a wireless signal, etc., for alerting the rear vehicle to brake. The visual signal, the infrared signal, the acoustic signal, the wireless signal, etc., may be sent through corresponding hardware units of the current vehicle. As an example, the rear vehicle brake alerting unit may be a brake signal light of the current vehicle. The brake signal light may be existing lights on the vehicle for indicating that the brake pedal is depressed, e.g., conventional brake signal lights at both ends of the rear or a high-mounted brake signal light in the upper middle of the rear. According to the embodiments of the present disclosure, a brake signal light on a vehicle may also be used to issue a rear vehicle brake alert to a vehicle behind it. Preferably, the rear vehicle brake alerting unit may be issued through a brake alerting light of the current vehicle. The brake alerting light may be a light added at the rear of the vehicle, and specially configured to issue a rear vehicle brake alert to a vehicle behind it. The shape of the brake alerting light, or the characteristics of the light it emits, may be significantly different from the brake signal light, so that the rear vehicle brake alert it emits may be easily detected by the rear vehicle.

If it is determined that the rear collision probability is less than or equal to the rear collision threshold at 212, the process 200 may return to 206, where a rear input may continue to be obtained. In this case, the current vehicle may not issue the rear vehicle brake alert to the rear vehicle, to avoid disturbing the rear vehicle.

It can be seen from the steps 212 to 214, if the rear collision probability is greater than the predetermined rear collision threshold, i.e., if the rear vehicle is likely to collide or rear-end the current vehicle, the rear vehicle brake alert may be issued to the rear vehicle; and if the rear collision probability is not greater than the predetermined rear collision threshold, i.e., if the rear vehicle is not likely to collide or rear-end the current vehicle, the rear vehicle brake alert will not be issued to the rear vehicle. That is, the rear vehicle brake alert may be issued to the rear vehicle only when the rear vehicle is likely to collide or rear-end the current vehicle. In the above manner, an accurate rear vehicle brake alert may be passed to a rear vehicle. Additionally, if each vehicle may determine, based on its rear context, whether or not to issue a rear vehicle brake alert to a vehicle behind it, accurate rear vehicle brake alerts may be passed in sequence to the rear, thereby significantly reducing the risk of collisions or rear-end collisions on roads.

If at 204 it is detected that the front vehicle is braking or is about to brake, the process 200 may further proceed to 216. At 216, a front context of the current vehicle may be obtained through analyzing the front input. The front context of the current vehicle may include, e.g., a distance between the front vehicle and the current vehicle, a relative speed of the front vehicle relative to the current vehicle, a road condition, a weather condition, a vehicle type of the front vehicle, etc. The front context may be obtained in a known way. For example, the distance between the front vehicle and the current vehicle may be obtained through a known inter-vehicle ranging technology. For the relative speed of the front vehicle relative to the current vehicle, a change of the distance between the front vehicle and the current vehicle within a predetermined time may be estimated first, and then the relative speed of the front vehicle relative to the current vehicle may be calculated based on the change and the predetermined time. The road condition, the weather condition, the vehicle type of the front vehicle, etc., may be obtained through a known image classification technology.

At 218, a front collision probability of the current vehicle colliding with the front vehicle may be predicted based on the front context of the current vehicle and an assumption that the front vehicle and the current vehicle brake. The front collision probability may be normalized to a number within a range [0, 1], A greater front collision probability means that there is a higher probability of the current vehicle colliding with the front vehicle, i.e., the current vehicle will more likely rear-end the front vehicle. The front collision probability may be predicted in a similar manner as the rear collision probability.

At 220, it may be determined whether the front collision probability is greater than a front collision threshold.

If at 220 it is determined that the front collision probability is greater than the front collision threshold, a current vehicle braking command may be sent within the current vehicle. It may be determined how to send the current vehicle braking command within the current vehicle according to a vehicle type of the current vehicle. The vehicle type of the current vehicle may include, e.g., a human-driven vehicle, an autonomous vehicle, etc.

The process 200 may proceed to 222. At 222, it may be determined whether the current vehicle is a human-driven vehicle.

If it is determined at 222 that the current vehicle is the human-driven vehicle, the process 200 may proceed to 224. At 224, the current vehicle brake command may be sent to a current vehicle brake alerting unit in the current vehicle, so that the current vehicle brake alerting unit issues a current vehicle brake alert. The current vehicle brake alert may alert a driver of the current vehicle to implement a braking operation as soon as possible, such as an emergency braking operation. The current vehicle brake alert may include, e.g., a visual signal, an acoustic signal, a haptic signal, etc. Preferably, the braking degree of the current vehicle brake alert issued by the current vehicle braking unit may be associated with the front collision probability. The greater the front collision probability, the greater the intensity of the current vehicle brake alert may be, so as to cause the driver to perform a braking operation with a greater braking degree. For example, when the current vehicle brake alert is an acoustic signal, if the front collision probability is a large value greater than the front collision threshold, the acoustic signal may be urgent; and if the front collision probability is a small value greater than the front collision threshold, the acoustic signal may be flat. If it is determined at 222 that the current vehicle is not a human-driven vehicle, the current vehicle may be an autonomous vehicle. In this case, the process 200 may proceed to 226. At 226, a current vehicle braking command may be sent at least to an automatic braking unit in the current vehicle, so that the automatic braking unit implements an automatic braking operation. Preferably, the braking degree of the automatic braking operation implemented by the automatic braking unit may be associated with the front collision probability. The greater the front collision probability, the greater the braking degree of the automatic braking operation may be, so as to cause the current vehicle to stop as soon as possible. Optionally, if the current vehicle is the autonomous vehicle, the current vehicle braking command may also be sent to the current vehicle brake alerting unit in the current vehicle, so that the driver in the current vehicle may also know that the current vehicle is about to be automatically braked, and may take over the vehicle and take appropriate actions if necessary.

If it is determined at 220 that the front collision probability is less than or equal to the front collision threshold, the process 200 may return to 202, where a front input continues to be obtained. In this case, the current vehicle may continue to drive forward for a certain distance, and may send the current vehicle braking command within the current vehicle only when the front collision probability is greater than the front collision threshold, so that the automatic driving unit or the driver of the current vehicle takes a braking operation. If at this point, the rear vehicle may collide or rear-end the current vehicle, and the rear vehicle is able to immediately perform emergency braking when receiving a rear vehicle brake alert, and the current vehicle continues to drive for a certain distance on the premise of ensuring that it will not rear-end the front vehicle, the distance between the rear vehicle and the current vehicle may be increased. In this way, the risk of the rear vehicle rear-ending the current vehicle may be reduced.

It should be appreciated that the process for brake alerting based on context detection described above in conjunction with FIG.2 is merely exemplary. Depending on actual application requirements, the steps in the process for brake alerting based on context detection may be replaced or modified in any manner, and the process may include more or fewer steps. Additionally, the specific order or hierarchy of the steps in the process 200 is merely exemplary, and the process for brake alerting based on context detection may be performed in an order different from the described one.

FIG.4 illustrates an exemplary process 400 for predicting a rear collision probability of a rear vehicle colliding with a current vehicle according to an embodiment of the present disclosure. The process 400 may correspond to the step 210 in FIG.2.

At 402, a time for a rear vehicle to collide with a current vehicle may be estimated based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake. The rear context may include, e.g., a distance between the rear vehicle and the current vehicle, a relative speed of the rear vehicle relative to the current vehicle, a road condition, a weather condition, a vehicle type of the rear vehicle, etc.

It is assumed that the time when the rear vehicle and the current vehicle start braking is an initial time 0, and the time for the rear vehicle to collide with the current vehicle is t _b . That is, the time for the rear vehicle to collide with the current vehicle may be denoted as t _b . During a period from the initial time 0 to the time t _b , the journey traveled by the rear vehicle and the journey 5 _c (tj,) traveled by the current vehicle may have the following relationship: where L _b is an initial distance between the rear vehicle and the current vehicle, V _b is an initial speed of the rear vehicle, V _c is an initial speed of the current vehicle, and t _d is a braking response time of the rear vehicle and the current vehicle. Herein, for simplicity, it is assumed that the rear vehicle and the current vehicle have the same braking response time. The braking response time t _d may be an empirical value, e.g., 0.4 seconds or 0.5 seconds.

The journey traveled by the rear vehicle from the initial time 0 to the time t _b may be calculated, e.g., through the following equation: where a _b is a braking acceleration of the rear vehicle.

The braking acceleration a _b of the rear vehicle is generated by a braking force f _b . The braking force f _b may be expressed, e.g., through the following equation: fb = -b * G _b = g _b * m _b * g (3) where g _b is a braking force coefficient of the rear vehicle, m _b is the mass of the rear vehicle, and g is a gravitational acceleration. The braking force coefficient g _b may be related to a road condition, a weather condition, a vehicle type of the rear vehicle, etc. The weather condition affects the road condition, e.g., how slippery the road is. The road condition affects the friction between the vehicle and the road surface, which in turn affects the braking force coefficient g _b . A vehicle type of a vehicle is related to the maximum allowable braking force of the vehicle, thereby affects the braking force coefficient g _b . For example, the maximum allowable braking force of a heavy-duty truck may be smaller than that of a light-duty car, which will make the braking distance of the heavy-duty truck greater than that of the light-duty car under the same condition.

Based on the equation (3), we can get:

Combining the equation (2) and the equation (4), we can get:

The journey S _c (tj,) traveled by the current vehicle from the initial time 0 to the time t _b may be calculated similarly, as follows:

Combining the equation (1), the equation (5) and the equation (6), we can get:

The equation (7) is a quadratic equation of one variable about t _b . This equation may be solved through known a solving method for the quadratic equation of one variable. For example, the equation (7) may have the following two solutions:

These two solutions may be a positive solution and a negative solution, respectively. The positive solution of the two solutions may be taken as the time t _b when the rear vehicle collides with the current vehicle.

It may be seen from the equation (8) that when the braking response time t _d is constant, the time t _b when the rear vehicle collides with the current vehicle may be related to the relative speed (V _b ~ ) °f the rear vehicle relative to the current vehicle, the difference q _b —q _c ) between the braking force coefficient q _b of the rear vehicle and the braking force coefficient _c of the current vehicle, and the initial distance L _b between the rear vehicle and the current vehicle.

Alternatively, in order to optimize the computational cost or improve the running speed, it may be assumed q _b = q _c . In this case, the equation (7) may be simplified as:

Vj, * t _b = V _c * t _b + L _b - (V _b - V _c ) * t _d (9)

Therefore, it may be solved:

It may be seen from the equation (10) that when the braking response time t _d is constant, the time t _b when the rear vehicle collides with the current vehicle may be related to the relative speed (Vb ~ ) °f the rear vehicle relative to the current vehicle and the initial distance L _b between the rear vehicle and the current vehicle.

After estimating the time t _b when the rear vehicle collides with the current vehicle, a rear collision probability may be calculated based on the estimated time t _b . Preferably, the rear collision probability may be normalized to a value within a range [0, 1], In an implementation, a normalized time may be obtained first, and then a normalized rear collision probability may be obtained based on the normalized time. For example, at 404, a normalized time t _b may be obtained through performing a normalization operation on the estimated time t _b . The normalized time t _b may be a value within the range [0, 1], Various known normalization methods may be employed to perform the normalization operation on the estimated time t _b . As an example, a normalization method may be a Min-Max Normalization method. Accordingly, the normalized time t _b may be calculated, e.g., through the following equation: 1 ¹¹ ) where t _min is the minimum value of the time t _b , and t _max is the maximum value of the time t _b . Theoretically, the time t _b may be any number from 0 to positive infinity. That is, t _min may be 0. t _max may be set to a predetermined value, e.g., 30 seconds. This means that if the rear vehicle does not collide with the current vehicle within 30 seconds, it may be considered that the rear vehicle will not collide with the current vehicle.

In the case where t _min is 0, the equation (11) may be further evolved as:

After the normalized time t _b ' is obtained, at 406, a rear collision probability P _b of the rear vehicle colliding with the current vehicle may be calculated based on the normalized time t _b ' . The rear collision probability P _b may be calculated, e.g., through the following equation: t _b is normalized to a value within the range [0, 1], Accordingly, P _b is also normalized to a value within the range [0, 1], In a case where t _max is constant, the smaller the value of t _b is, the closer P _b will be to 1, that is, the greater the rear collision probability will be; while the greater the value of t _b is, the closer P _b will be to 1, that is, the smaller the rear collision probability will be.

It should be appreciated that the process for predicting the rear collision probability described above in conjunction with FIG.4 is merely exemplary. Depending on actual application requirements, the steps in the process for predicting the rear collision probability may be replaced or modified in any manner, and the process may include more or fewer steps. For example, in addition to employing the max-min normalization method to perform the normalization operation on the estimated time t _b , any other normalization method may also be employed to perform the normalization operation on the estimated time t _b . Additionally, after obtaining the normalized time t _b , in addition to calculating the rear collision probability P _b through the equation (13), other methods may also be employed to calculate the rear collision probability P _b .

A front collision probability Py of a current vehicle colliding with a front vehicle may be predicted through a process similar to the process 400. For example, a time tf when the current vehicle collides with the front vehicle may be estimated based on a front context of the current vehicle and an assumption that the current vehicle and the front vehicle brake. Subsequently, a normalized time t may be obtained through performing a normalization operation on the estimated time tf. Next, a front collision probability Pf of the current vehicle colliding with the front vehicle may be calculated based on the normalized time t .

FIGs.5A to 5G illustrate examples 500a to 500g of brake alerting based on context detection according to embodiments of the present disclosure.

In the example 500a in FIG.5A, the weather condition and the road condition are normal. A front vehicle 504a in front of a current vehicle 502a is performing an emergency braking. The distance between the front vehicle 504a and the current vehicle 502a is long. A rear vehicle 506a behind the current vehicle 502a is a truck, and the distance between the rear vehicle 506a and the current vehicle 502a is short. A rear collision probability of the rear vehicle 506a colliding with the current vehicle 502a, predicted through the process 200 in FIG.2, may be greater than a rear collision threshold. In this case, the current vehicle 502a may issue a rear vehicle brake alert to the rear vehicle 506a, to alert the rear vehicle 506a to brake immediately. For example, the current vehicle 502a may have its brake signal light or brake alerting light illuminated immediately. Meanwhile, a front collision probability of the current vehicle 502a colliding with the front vehicle 504a, predicted through the process 200 in FIG.2, may be less than a front collision threshold. In this case, the current vehicle 502a may not take a braking operation. In this case, the current vehicle 502a may continue to drive forward for a certain distance, and may take a braking operation only when the front collision probability is greater than the front collision threshold. If the rear vehicle 506a is able to immediately perform emergency braking when detecting the rear vehicle brake alert issued by the current vehicle 502a, and the current vehicle 502a continues to drive for a certain distance on the premise of ensuring that it will not rear-end the front vehicle 504a, the distance between the rear vehicle 506a and the current vehicle 502a may be increased. In this way, the risk of the rear vehicle 506a rear-ending the current vehicle 502a may be reduced.

In the example 500b in FIG.5B, the weather condition and the road condition are normal. A front vehicle 504b in front of a current vehicle 502b is performing an emergency braking. The distance between a front vehicle 504b and the current vehicle 502b is moderate. The distance between the rear vehicle 506b behind the current vehicle 502b and the current vehicle 502b is moderate. A rear collision probability of the rear vehicle 506b colliding with the current vehicle 502b, predicted through the process 200 in FIG.2, may be less than a rear collision threshold. In this case, the current vehicle 502b may not issue a rear vehicle brake alert to the rear vehicle 506b, to avoid disturbing the rear vehicle 506b. Meanwhile, a front collision probability of the current vehicle 502b colliding with the front vehicle 504b, predicted through the process 200 in FIG.2, may be less than a front collision threshold. In this case, the current vehicle 502b may not take a braking operation. In this case, the current vehicle 502b may continue to drive forward for a certain distance, and may take a braking operation only when the front collision probability is greater than the front collision threshold.

In the example 500c in FIG.5C, it is raining and the road is wet and slippery. A front vehicle 504c in front of a current vehicle 502c is performing an emergency braking. The distance between the front vehicle 504c and the current vehicle 502c is long. The distance between a rear vehicle 506c behind the current vehicle 502c and the current vehicle 502c is short. A rear collision probability of the rear vehicle 506c colliding with the current vehicle 502c, predicted through the process 200 in FIG.2, may be greater than a rear collision threshold. In this case, the current vehicle 502c may issue a rear vehicle brake alert to the rear vehicle 506c. For example, the current vehicle 502c may have its brake signal light or brake alerting light illuminated immediately, to alert the rear vehicle 506c to brake immediately. Meanwhile, a front collision probability of the current vehicle 502c colliding with the front vehicle 504c, predicted through the process 200 in FIG.2, may be less than a front collision threshold. In this case, the current vehicle 502c may not take a braking operation. In this case, the current vehicle 502c may continue to drive forward for a certain distance, and may take a braking operation only when the front collision probability is greater than the front collision threshold. If the rear vehicle 506c is able to immediately perform emergency braking when detecting the rear vehicle brake alert issued by the current vehicle 502c, and the current vehicle 502c continues to drive for a certain distance on the premise of ensuring that it will not rear-end the front vehicle 504c, the distance between the rear vehicle 506c and the current vehicle 502c may be increased. In this way, the risk of the rear vehicle 506c rear-ending the current vehicle 502c may be reduced.

In the example 500d in FIG.5D, the weather condition and the road condition are normal. A front vehicle 504d in front of a current vehicle 502d is performing an emergency braking. The distance between the front vehicle 504d and the current vehicle 502d is short. The distance between a rear vehicle 506d behind the current vehicle 502d and the current vehicle 502d is long. A rear collision probability of the rear vehicle 506d colliding with the current vehicle 502d, predicted through the process 200 in FIG.2, may be less than a rear collision threshold. In this case, the current vehicle 502d may not issue a rear vehicle brake alert to the rear vehicle 506d, to avoid disturbing the rear vehicle 506d. Meanwhile, a front collision probability of the current vehicle 502d colliding with the front vehicle 504d, predicted through the process 200 in FIG.2, may be greater than a front collision threshold. In this case, the current vehicle 502d may immediately take an emergency braking operation, to avoid colliding with the front vehicle 504d. In the example 500e in FIG.5E, the weather condition and the road condition are normal. A front vehicle 504e in front of a current vehicle 502e is performing a braking but not an emergency braking. The distance between the front vehicle 504e and the front vehicle 502e is moderate. A rear vehicle 506e behind the current vehicle 502e is a truck, and the distance between the rear vehicle 506e and the current vehicle 502e is moderate. A rear collision probability of the rear vehicle 506e colliding with the current vehicle 502e, predicted through the process 200 in FIG.2, may be less than a rear collision threshold. In this case, the current vehicle 502e may not issue a rear vehicle brake alert to the rear vehicle 506e, to avoid disturbing the rear vehicle 506e. Meanwhile, a front collision probability of the current vehicle 502e colliding with the front vehicle 504e, predicted through the process 200 in FIG.2, may be less than a front collision threshold. In this case, the current vehicle 502e may not take a braking operation. In this case, the current vehicle 502e may continue to drive forward for a certain distance, and may take a braking operation only when the front collision probability is greater than the front collision threshold.

In the example 500f in FIG.5F, the weather condition and the road condition are normal. A front vehicle 504f in front of a current vehicle 502f has not braked yet, but its brake alerting light is illuminating. The distance between the front vehicle 504f and the current vehicle 502f is moderate. A rear vehicle 506f behind the current vehicle 502f is a truck, and the distance between the rear vehicle 506f and the current vehicle 502f is short. A rear collision probability of the rear vehicle 506f colliding with the current vehicle 502f, predicted through the process 200 in FIG.2, may be greater than a rear collision threshold. In this case, the current vehicle 502f may issue a rear vehicle brake alert to the rear vehicle 506f. For example, the current vehicle 502f may have its brake signal light or brake alerting light illuminated immediately, to alert the rear vehicle 506f to brake immediately. Meanwhile, the front collision probability of the current vehicle 502f colliding with the front vehicle 504f, predicted through the process 200 in FIG.2, may be less than a front collision threshold. In this case, the current vehicle 502f may not take a braking operation. In this case, the current vehicle 502f may continue to drive forward for a certain distance, and may take a braking operation only when the front collision probability is greater than the front collision threshold. If the rear vehicle 506f is able to immediately perform emergency braking when detecting the rear vehicle brake alert issued by the current vehicle 502f, and the current vehicle 502f continues to drive for a certain distance on the premise of ensuring that it will not rear-end the front vehicle 504f, the distance between the rear vehicle 506f and the current vehicle 502f may be increased. In this way, the risk of the rear vehicle 506f rear-ending the current vehicle 502f may be reduced.

In the example 500g in FIG.5G, the weather condition and the road condition are normal. A front vehicle 504g in front of a current vehicle 502g has not braked yet, but its brake alerting light is illuminating. The distance between the front vehicle 504g and the current vehicle 502g is short. The distance between a rear vehicle 506g behind the current vehicle 502g and the current vehicle 502g is long. A rear collision probability of the rear vehicle 506g colliding with the current vehicle 502g, predicted through the process 200 in FIG.2, may be less than a rear collision threshold. In this case, the current vehicle 502g may not issue a rear vehicle brake alert to the rear vehicle 506g, to avoid disturbing the rear vehicle 506g. Meanwhile, a front collision probability of the current vehicle 502g colliding with the front vehicle 504g, predicted through the process 200 in FIG.2, may be greater than a front collision threshold. In this case, the current vehicle 502g may take an emergency braking operation, to avoid colliding with the front vehicle 504g.

The brake alerting based on context detection according to the embodiments of the present disclosure is described above. When a current vehicle detects that a front vehicle is braking or is about to brake, it may issue a rear vehicle brake alert to a rear vehicle. The current vehicle may be in the same lane as the front vehicle and the rear vehicle. Optionally, the current vehicle may also detect whether a side front vehicle in an adjacent lane is braking or is about to brake, and issue a brake alert to a side rear vehicle in the adjacent lane when detects the side front vehicle in the adjacent lane is braking or is about to brake. The brake alert issued to the side rear vehicle may be different from the brake alert issued to the rear vehicle, e.g., different lights may be illuminated, differently encoded wireless signals may be sent, etc.

FIG.6 is a flowchart of an exemplary method 600 for brake alerting based on context detection according to an embodiment of the present disclosure.

At 610, it may be detected whether a front vehicle in front of a current vehicle is braking or is about to brake.

At 620, in response to detecting that the front vehicle is braking or is about to brake, a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle may be predicted based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake.

At 630, it may be determined whether the rear collision probability is greater than a rear collision threshold.

At 640, in response to determining that the rear collision probability is greater than the rear collision threshold, a rear vehicle brake alert may be issued to the rear vehicle.

In an implementation, the detecting whether a front vehicle is braking or is about to brake may comprise: detecting whether the front vehicle issues a brake alert; and/or detecting whether the drop in an absolute speed of the front vehicle exceeds a drop threshold.

The detecting whether the front vehicle issues a brake alert may comprise: detecting whether the front vehicle issues at least one of a visual signal, an infrared signal, an acoustic signal and a wireless signal for alerting the current vehicle to brake.

The visual signal may be issued through a brake signal light or a brake alerting light of the front vehicle.

The absolute speed of the front vehicle may be calculated through: estimating a change in a distance between the front vehicle and the current vehicle over a predetermined time; calculating a relative speed of the front vehicle relative to the current vehicle based on the change and the predetermined time; and calculating the absolute speed of the front vehicle based on the relative speed and the absolute speed of the current vehicle.

In an implementation, the predicting a rear collision probability may comprise: estimating a time for the rear vehicle to collide with the current vehicle based on the rear context and the assumption that the current vehicle and the rear vehicle brake; obtaining a normalized time through performing a normalization operation on the estimated time; and calculating the rear collision probability based on the normalized time.

In an implementation, the rear context may include: a relative speed of the rear vehicle relative to the current vehicle and a distance between the rear vehicle and the current vehicle.

In an implementation, the issuing a rear vehicle brake alert may comprise: issuing, to the rear vehicle, at least one of a visual signal, an infrared signal, an acoustic signal and a wireless signal for alerting the rear vehicle to brake.

The visual signal may be to be issued through a brake signal light or a brake alerting light of the current vehicle.

In an implementation, the method 600 may further comprise: in response to detecting that the front vehicle is braking or is about to brake, predicting a front collision probability of the current vehicle colliding with the front vehicle based on a front context of the current vehicle and an assumption that the front vehicle and the current vehicle brake; determining whether the front collision probability is greater than a front collision threshold; and in response to determining that the front collision probability is greater than the front collision threshold, sending a current vehicle braking command within the current vehicle.

The front context may include: a relative speed of the front vehicle relative to the current vehicle and a distance between the front vehicle and the current vehicle.

The current vehicle may be a human-driven vehicle. The sending a current vehicle brake command may comprise: sending the current vehicle brake command to a current vehicle brake alerting unit in the current vehicle, so that the current vehicle brake alerting unit issues a current vehicle brake alert.

The current vehicle brake alert may include at least one of a visual signal, an acoustic signal, and a haptic signal. The intensity of the current vehicle brake alert may be associated with the front collision probability.

The current vehicle may be an autonomous vehicle. The sending a current vehicle braking command may comprise: sending the current vehicle braking command at least to an automatic braking unit in the current vehicle, so that the automatic braking unit implements an automatic braking operation.

The braking degree of the automatic braking operation may be associated with the front collision probability.

The rear context may further include at least one of a road condition, a weather condition, a vehicle type of the rear vehicle. The front context may also include at least one of a road condition, a weather condition, and a vehicle type of the front vehicle.

It should be appreciated that the method 600 may further comprise any other steps/process for brake alerting based on context detection according to the embodiments of the present disclosure as mentioned above.

FIG.7 illustrates an exemplary system 700 forbrake alerting based on context detection according to an embodiment of the present disclosure.

The system 700 may comprise: a processing unit 710, configured to: detect whether a front vehicle in front of a current vehicle is braking or is about to brake, in response to detecting that the front vehicle is braking or is about to brake, predict a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake, determine whether the rear collision probability is greater than a rear collision threshold, and in response to determining that the rear collision probability is greater than the rear collision threshold, send a rear vehicle brake alerting command to a rear vehicle brake alerting unit; and a rear vehicle brake alerting unit 720, configured to: in response to receiving the rear vehicle brake alerting command from the processing unit, issue a rear vehicle brake alert to the rear vehicle.

The processing unit 710 may further be configured to: in response to detecting that the front vehicle is braking or is about to brake, predict a front collision probability of the current vehicle colliding with the front vehicle based on a front context of the current vehicle and an assumption that the front vehicle and the current vehicle brake; determine whether the front collision probability is greater than a front collision threshold; and in response to determining that the front collision probability is greater than the front collision threshold, send a current vehicle braking command within the current vehicle. The system 700 may further comprise: a current vehicle brake alerting unit 730, configured to in response to receiving the current vehicle braking command from the processing unit, issue a current vehicle brake alert; and/or an automatic braking unit 740, configured to in response to receiving the current vehicle braking command from the processing unit, implement an automatic braking operation.

It should be appreciated that the system 700 may further comprise other components for performing brake alerting based on context detection according to the embodiments of the present disclosure as mentioned above.

FIG.8 illustrates an exemplary apparatus 800 for brake alerting based on context detection according to an embodiment of the present disclosure.

The apparatus 800 may comprise at least one processor 810 and a memory 820 storing computerexecutable instructions. The computer-executable instructions, when executed, may cause the at least one processor 810 to: detect whether a front vehicle in front of a current vehicle is braking or is about to brake, in response to detecting that the front vehicle is braking or is about to brake, predict a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake, determine whether the rear collision probability is greater than a rear collision threshold, and in response to determining that the rear collision probability is greater than the rear collision threshold, issue a rear vehicle brake alert to the rear vehicle.

It should be appreciated that the processor 810 may further perform any other steps/processes for brake alerting based on context detection according to the embodiments of the present disclosure as mentioned above.

The embodiments of the present disclosure propose a computer program product for brake alerting based on context detection, comprising a computer program that is executed by at least one processor for: detecting whether a front vehicle in front of a current vehicle is braking or is about to brake; in response to detecting that the front vehicle is braking or is about to brake, predicting a rear collision probability of a rear vehicle behind the current vehicle colliding with the current vehicle based on a rear context of the current vehicle and an assumption that the current vehicle and the rear vehicle brake; determining whether the rear collision probability is greater than a rear collision threshold; and in response to determining that the rear collision probability is greater than the rear collision threshold, issuing a rear vehicle brake alert to the rear vehicle. Additionally, the computer program may further be performed for implementing any other steps/processes for brake alerting based on context detection according to the embodiments of the present disclosure as mentioned above.

The embodiments of the present disclosure may be embodied in a non-transitory computer- readable medium. The non-transitory computer readable medium may comprise instructions that, when executed, cause one or more processors to perform any operation of the method for brake alerting based on context detection according to the embodiments of the present disclosure as mentioned above.

It should be appreciated that all the operations in the methods described above are merely exemplary, and the present disclosure is not limited to any operations in the methods or sequence orders of these operations, and should cover all other equivalents under the same or similar concepts. In addition, the articles “a” and “an” as used in this specification and the appended claims should generally be construed to mean “one” or “one or more” unless specified otherwise or clear from the context to be directed to a singular form.

It should also be appreciated that all the modules in the apparatuses described above may be implemented in various approaches. These modules may be implemented as hardware, software, or a combination thereof. Moreover, any of these modules may be further functionally divided into sub-modules or combined together.

Processors have been described in connection with various apparatuses and methods. These processors may be implemented using electronic hardware, computer software, or any combination thereof. Whether such processors are implemented as hardware or software will depend upon the particular application and overall design constraints imposed on the system. By way of example, a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be implemented with a microprocessor, microcontroller, digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic device (PLD), a state machine, gated logic, discrete hardware circuits, and other suitable processing components configured for performing the various functions described throughout the present disclosure. The functionality of a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be implemented with software being executed by a microprocessor, microcontroller, DSP, or other suitable platform.

Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, threads of execution, procedures, functions, etc. The software may reside on a computer-readable medium. A computer-readable medium may include, by way of example, memory such as a magnetic storage device (e.g., hard disk, floppy disk, magnetic strip), an optical disk, a smart card, a flash memory device, random access memory (RAM), read only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), a register, or a removable disk. Although memory is shown separate from the processors in the various aspects presented throughout the present disclosure, the memory may be internal to the processors, e.g., cache or register.

The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein. All structural and functional equivalents to the elements of the various aspects described throughout the present disclosure that are known or later come to be known to those of ordinary skilled in the art are expressly incorporated herein and intended to be encompassed by the claims.